Twitter/X Conversion Window Mismatch
One or more X Pixel events on this account are reporting on a conversion window that does not line up with the window set at the campaign level, or the window is shorter than the sales cycle you actually run. The result is reporting that cuts off valid conversions, which then feeds back into optimisation and makes campaigns look worse than they are.
Why It Matters
X conversion tracking supports post-click windows of 1, 7, 14, 30, and 90 days, plus view-through windows of 1 and 7 days. When an event-level window is shorter than the campaign-level window, the campaign sees fewer conversions than the event recorded, and your ROAS column reads low. When the event-level window is longer than the campaign window, the opposite happens and the totals do not reconcile across views. The second variant is a window that is honest about itself but mismatched to the business. A 7-day post-click window on a B2B SaaS account with a 45-day median sales cycle truncates the majority of real conversions. Optimisation runs against the short tail it can see, learns from the wrong signal, and bids accordingly. The account does not look broken in Events Manager because the events that do land look clean. The damage is in everything that never got attributed. Mismatches usually appear after a window change at one level (event or campaign) that was not mirrored at the other, or after a campaign was duplicated from a template that predates a window policy update.
How To Fix It
- In X Ads Events Manager, list every active conversion event and note the attribution window on each.
- Open every campaign that optimises against those events and compare the campaign window to the event window. Resolve mismatches by setting both to the same value.
- Pick the window from the real sales cycle, not from a template. Ecommerce checkout typically lives at 1 or 7 days post-click. Considered B2B purchases usually need 30 to 90 days.
- Document the chosen window per event in the measurement plan so future campaign duplications inherit the right setting.
- After changing windows, allow at least one full cycle of new data before judging the impact, since historical conversions are not reattributed.
Example
Event Purchase: 7-day post-click attribution window
Campaign optimising Purchase: 1-day post-click attribution window
Result: campaign reports fewer Purchases than Events Manager, ROAS reads low.Your X Pixel events are configured with conversion windows that do not align with the matching campaign settings, or the windows are shorter than the real sales cycle for this business. Per X Ads Help Center documentation on conversion tracking for websites, attribution windows are configurable per event and per campaign across the supported post-click options of 1, 7, 14, 30, and 90 days plus view-through options of 1 and 7 days, and reporting only counts conversions that fall inside the configured window. A mismatch between event and campaign settings produces reports that fail to reconcile across views, and a window that is shorter than the sales cycle truncates valid conversions and skews optimisation toward the short tail that the platform can see. Fix: align the event-level and campaign-level windows for every conversion event, pick the window from the real sales cycle rather than the template default, and document the chosen window per event so future campaign duplications inherit it. Source: business.twitter.com/en/help/campaign-measurement-and-analytics/conversion-tracking-for-websites.html.
Drop this paragraph into your client deliverable. Sources back to the canonical platform documentation linked below.
References
Audit your own files for this check
AdLint runs this check (and 177 others) against your GTM, Google Ads, Meta, TikTok, LinkedIn, Pinterest, Twitter/X, and Snapchat exports. Everything stays in your browser. No uploads, no accounts.
Run a free audit